[LLDB][NFC]Fix stack-use-after free bug.

[oontvoo]

Details: detailed_command_telemetry (bool) and command_id (int) could already be freed when the dispatcher's dtor runs. So we should just copy them into the lambda since they are cheap.

[llvmbot]

@llvm/pr-subscribers-lldb

Author: Vy Nguyen (oontvoo)

Changes

Details: detailed_command_telemetry (bool) and command_id (int) could already be freed when the dispatcher's dtor runs. So we should just copy them into the lambda since they are cheap.

---

Full diff: https://github.com/llvm/llvm-project/pull/134296.diff

1 Files Affected:

• (modified) lldb/source/Interpreter/CommandInterpreter.cpp (+3-1)

diff --git a/lldb/source/Interpreter/CommandInterpreter.cpp b/lldb/source/Interpreter/CommandInterpreter.cpp
index 949b1191c28f0..112d2f20fda41 100644
--- a/lldb/source/Interpreter/CommandInterpreter.cpp
+++ b/lldb/source/Interpreter/CommandInterpreter.cpp
@@ -1918,7 +1918,9 @@ bool CommandInterpreter::HandleCommand(const char *command_line,
     // Those will be collected by the on-exit-callback.
   });
 
-  helper.DispatchOnExit([&](lldb_private::telemetry::CommandInfo *info) {
+  helper.DispatchOnExit([&cmd_obj, &parsed_command_args, &result,
+                         detailed_command_telemetry, command_id](
+                            lldb_private::telemetry::CommandInfo *info) {
     // TODO: this is logging the time the command-handler finishes.
     // But we may want a finer-grain durations too?
     // (ie., the execute_time recorded below?)

[oontvoo]

I think this is a trivial enough fix - i'm going to just merge it.